Compare XPUs

Select up to 5 XPUs to compare side-by-side

Select XPUs to Compare

Clear all (5)Jump to results

Filter by Vendor

Showing 72 XPUs • 5 selected

Alibaba

Hanguang 800

AMD

MI100

23.1 TFLOPs

AMD

MI210

181 TFLOPs

AMD

MI250X

383 TFLOPs

AMD

MI300A

980.6 TFLOPs

AMD

MI300X

1,307 TFLOPs

AMD

MI325X

1,400 TFLOPs

AMD

MI350X

2,100 TFLOPs

AMD

Radeon PRO W7900

122 TFLOPs

AMD

Radeon RX 7900 XT

104 TFLOPs

AMD

Radeon RX 7900 XTX

122 TFLOPs

AWS

Inferentia2

190 TFLOPs

AWS

Trainium

190 TFLOPs

AWS

Trainium2

680 TFLOPs

Baidu

Kunlun II

Biren Technology

BR100

Cambricon

MLU370

256 TFLOPs

Cerebras

WSE-3

Enflame Technology

CloudBlazer T20

Etched

Sohu

10,000 TFLOPs

FuriosaAI

Warboy

Google

TPU v4

275 TFLOPs

Google

TPU v5e

197 TFLOPs

Google

TPU v5p

459 TFLOPs

Google

TPU v6e (Trillium)

918 TFLOPs

Graphcore

Bow IPU

Graphcore

IPU-M2000

Groq

LPU Inference Engine

Huawei

Ascend 910B

Iluvatar CoreX

BI-V150

300 TFLOPs

Intel

Data Center GPU Max 1100

177 TFLOPs

Intel

Data Center GPU Max 1550

419 TFLOPs

Intel Habana

Gaudi 2

432 TFLOPs

Intel Habana

Gaudi 3

1,835 TFLOPs

Multi-Metric Comparison

Relative performance across 5 key metrics (normalized to 100 = best in comparison)

Compute Performance (BF16)

Memory Capacity

Power Consumption

Power Efficiency

Specifications

Specification	NVIDIA L4	AMD MI100	AMD MI350X	Cerebras WSE-3	Baidu Kunlun II
Architecture	Ada Lovelace	CDNA	CDNA 4	Wafer Scale Engine	Kunlun Core
Form Factor	PCIe	PCIe	OAM	—	—
VRAM	24 GB	32 GB	288 GB	44 GB	32 GB
Memory Bandwidth	300 GB/s	1,228 GB/s	—	—	—
TFLOPs (FP32)	30.3	23.1	—	—	—
TFLOPs (FP16)	242	184.6	—	—	256
TFLOPs	121	23.1	2,100	—	—
TFLOPs (FP8)	485	—	—	—	—
TDP	72 W	300 W	850 W	23000 W	200 W
Launch Date	Mar 2023	Nov 2020	Jun 2025	Mar 2024	Aug 2021

Efficiency Metrics

Metric	L4	MI100	MI350X	WSE-3	Kunlun II
TFLOPs per Watt (FP32-eq)	0.84	0.08	1.24	—	—
Memory Bandwidth per GB	12.5 GB/s	38.4 GB/s	—	—	—

Performance Equivalence

How many units of each GPU are needed to match the performance of the others?

To match 1x NVIDIA L4

AMD MI100

Compute (FP32-eq)

2.62x

Need 2.62x MI100

FP32 Compute

1.31x

Need 1.31x MI100

VRAM

0.75x

MI100 has 1.33x more

Memory Bandwidth

0.24x

MI100 has 4.09x more

AMD MI350X

Compute (FP32-eq)

0.06x

MI350X is 17.36x faster

VRAM

0.08x

MI350X has 12.00x more

Cerebras WSE-3

VRAM

0.55x

WSE-3 has 1.83x more

Baidu Kunlun II

VRAM

0.75x

Kunlun II has 1.33x more

To match 1x AMD MI100

NVIDIA L4

Compute (FP32-eq)

0.38x

L4 is 2.62x faster

FP32 Compute

0.76x

L4 is 1.31x faster

VRAM

1.33x

Need 1.33x L4

Memory Bandwidth

4.09x

Need 4.09x L4

AMD MI350X

Compute (FP32-eq)

0.02x

MI350X is 45.45x faster

VRAM

0.11x

MI350X has 9.00x more

Cerebras WSE-3

VRAM

0.73x

WSE-3 has 1.38x more

Baidu Kunlun II

VRAM

1.00x

Kunlun II has 1.00x more

To match 1x AMD MI350X

NVIDIA L4

Compute (FP32-eq)

17.36x

Need 17.36x L4

VRAM

12.00x

Need 12.00x L4

AMD MI100

Compute (FP32-eq)

45.45x

Need 45.45x MI100

VRAM

9.00x

Need 9.00x MI100

Cerebras WSE-3

VRAM

6.55x

Need 6.55x WSE-3

Baidu Kunlun II

VRAM

9.00x

Need 9.00x Kunlun II

To match 1x Cerebras WSE-3

NVIDIA L4

VRAM

1.83x

Need 1.83x L4

AMD MI100

VRAM

1.38x

Need 1.38x MI100

AMD MI350X

VRAM

0.15x

MI350X has 6.55x more

Baidu Kunlun II

VRAM

1.38x

Need 1.38x Kunlun II

To match 1x Baidu Kunlun II

NVIDIA L4

VRAM

1.33x

Need 1.33x L4

AMD MI100

VRAM

1.00x

MI100 has 1.00x more

AMD MI350X

VRAM

0.11x

MI350X has 9.00x more

Cerebras WSE-3

VRAM

0.73x

WSE-3 has 1.38x more

Pricing

Price Type	L4	MI100	MI350X	WSE-3	Kunlun II
CAPEX (Street Price)	$4,000	—	—	—	—
OPEX (per hour)	$0.80/hr	—	—	—	—
Price per TFLOPs (FP32-eq)	$66	—	—	—	—